Incumbent presidents' NPV percentage compared to their approval ratings...

[twenty42]

2012 Obama
Approval rating: 50.0%
NPV percentage: 51.1%
Spread: +1.1%

2004 Bush
Approval rating: 49.1%
NPV percentage: 50.7%
Spread: +1.6%

1996 Clinton
Approval rating: 54.6%
NPV percentage: 49.2%
Spread: -5.4%

1992 Bush
Approval rating: 32.6%
NPV percentage: 37.4%
Spread: +4.8%

1984 Reagan
Approval rating: 57.9%
NPV percentage: 58.8%
Spread: +0.9%

1980 Carter
Approval rating: 37.9%
NPV percentage: 41.0%
Spread: +3.1%

1976 Ford
Approval rating: 43.6%
NPV percentage: 48.0%
Spread: +4.4%

1972 Nixon
Approval rating: 61.3%
NPV percentage: 60.7%
Spread: -0.6%

Averages since 1972
Approval rating: 48.4%
NPV percentage: 49.6%
Spread: +1.2%

So incumbents tend to slightly outperform their approval rating in most cases. Of course, the oddest case here is 1996 Clinton. Apparently, there were quite a few people who liked Clinton but still voted against him. It is possible that Perot threw a wrench into things, but GHWB still managed to outdo his approvals by almost five points in 1992 when Perot did eleven points better. It's also possible that there were quite a few voters who liked more than one candidate in 1996, but preferred Dole or Perot over Clinton. I can't imagine liking more than one candidate today.

I'm not assigning any predictive value to these numbers vis a vis 2020, as the numbers obviously show a wide range of possibilities. But it is worth noting that the average approval rating for an incumbent is 48.4%...not a far cry from Trump's current numbers.

[Kyle Rittenhouse is a Political Prisoner]

So given Trump's current approval, that'd predict 43% of the vote. Which almost certainly isn't enough to win unless there's a massive left wing third party presence.

[twenty42]

So given Trump's current approval, that'd predict 43% of the vote. Which almost certainly isn't enough to win unless there's a massive left wing third party presence.

If it is a 1976 or 1992 situation, that could put Trump at 47-48%. If it is a 1996 situation, he could wind up with 38-39%. Like I said, there has been a wide range of outcomes so it's hard to place any predictive value in the average.

Bear in mind too that presidential approval ratings tend to be higher in Year 4 than in Year 2.

[Thunder98]

It appears Incumbents tend to do better than there approve ratings most of the time.

[TheElectoralBoobyPrize]

In 1996, Clinton's percent of the two-party vote was actually almost his exact approval rating. Maybe most Perot voters approved....

Of course if you do 1992 this way, Bush Sr. outperformed his approval rating by even more.

[OSR stands with Israel]

So given Trump's current approval, that'd predict 43% of the vote. Which almost certainly isn't enough to win unless there's a massive left wing third party presence.

If it is a 1976 or 1992 situation, that could put Trump at 47-48%. If it is a 1996 situation, he could wind up with 38-39%. Like I said, there has been a wide range of outcomes so it's hard to place any predictive value in the average.

Bear in mind too that presidential approval ratings tend to be higher in Year 4 than in Year 2.

not for Carter or HW though